DESCRIPTION: Neurons in the brains of different mammals are almost indistinguishable in structure and function, but their numbers vary by more than three orders of magnitude. This variation has part of its origins in a set of undefined genes that control rates of cell proliferation and cell death in different pools of neural precursor cells. Some genes must have global effects and control the scale of the entire CNS. Other genes must have specific effects on specific cell populations. collectively, these genes are not only at the root of brain evolution, but they are very likely to be key genes that control the normal development of the brain. Major technical advances now make it practical for the first time to map genes responsible for complex neuronal traits. These advances include (1) a quadrupling of the density of the genetic map, (2) the ease of genotyping marker loci distributed across the entire genome, (3) biometric methods to map quantitative trait loci (QTLs), and (4) novel magnetic resonance imaging methods that can generate precise quantitative data on many different parts of the brain. The project proposed by the applicant will use these new tools to map putative genes that contribute to the regulation of cell number in various CNS structures. The applicant proposes three specific ways to approach this problem. In the first, he will analyze two sets of recombinant-inbred mouse strains (BXD and BXH) to determine the number and location of genes controlling retinal ganglion cell number in mouse. This effort has already yielded a locus on chromosome 11 (Rcn-1) that is responsible for the majority of the genetic differences in the two parental strains. In year three the applicant proposes to repeat this analysis on brains from F2 intercross progeny of (CAST/Ei x BALB/cJ)F1s. Part of this first aim will be to do a developmental analysis of the highest and lowest strains in order to determine whether the differences arise from cell death or cell division. The second aim will be to expand this study to the target of the retinal ganglion cells, namely the lateral geniculate nucleus. Counts of both source and target neurons will be made in large numbers of animals and the variations will be correlated with each other as well as with segregating loci in two diverse strains, BALB/cJ and C57BL/Ks as well as F2 intercross progeny of (CAST/Ei x BALB/cJ)F1s as well as various BXD recombinant inbred lines. The third specific aim will be similar in approach but instead of using cell counts of neurons in the visual pathway, volumetric measurements will be taken using a customized MRI radio frequency coil to analyze fixed brain tissue. The analysis will be done on different RI lines as well as mice from the (CAST/Ei x BALB/cJ)F1 intercross.